Search

Software effort estimation is a vital process in the software industry for successfully administering 5Ds of the software development life cycle (SDLC). The 5Ds stand for demand, development, direction, deployment, and designated cost of the software. Software development effort estimation (SDEE) is an effort prediction mechanism to calculate the effort for the development of the software product in order to minimize the challenges in the software field. Academics and practitioners are striving to identify which machine learning estimation technique yields more accurate results based on evaluation metrics, datasets, and other pertinent aspects. The feature selection techniques impact accuracy by selecting the main and relevant features in the dataset and eliminating the redundant and irrelevant features in the dataset. To achieve accurate estimations, the paper utilizes feature selection algorithms, along with various machine learning techniques, which predict the desired effort and the performance of the model has been measured in terms of prediction accuracy, R2 value, relative error, and mean absolute error. The datasets China and Maxwell are trained with the relevant features by applying feature selection algorithms, and estimation techniques are applied to predict the effort. The performance is compared with the regression models and feature selection techniques utilized by many authors previously. The result of the proposed methodology significantly gives the best performance with the combination of feature selection and estimation models than all regression models when applied alone, to both datasets. From the results, it is perceptible that random forest is performing well with the feature selection techniques and obtains the highest prediction accuracy of 99.33% with the China and 89.47% with the Maxwell datasets. [ABSTRACT FROM AUTHOR]

In this study, rice bran oil and sunn hemp fibre as reinforcement along with shellac and maleic anhydride were used to fabricate composite sheets. Rice bran oil was obtained from agricultural by-product, epoxidation of rice bran oil (ERBO) was carried, and further acrylation was carried to give acrylated epoxidized rice bran oil (AERBO) by chemical synthesis method. ERBO and AERBO were characterized in terms of iodine value conversion, oxirane value conversion, FTIR, and NMR. This synthesized AERBO was applied in composite fabrication along with maleic anhydride and shellac to study mechanical properties. The elongation at beak was highest about 40% for AERBO composite due to rubbery nature and 25% for maleaniated shellac (MS-AERBO) due to the stiff nature. The mechanical strength for AERBO in terms of tensile strength gave 7.9 MPa longitudinal 8.0 MPa transverse direction, and in the case of MS-AERBO values of 33.38 MPa longitudinal and 33. 31 MPa transverse, flexural strength gave values of 9.6 MPa for AERBO, and 26.06 MPa for MS-AERBO, and impact strength of 0.311 Izod and 1.100 Charpy for AERBO and MS-AERBO gave 1.15 and 3.14 for Izod and Charpy, and D shore hardness of 35 for AERBO, and 75 in the case of MS-AERBO composite. Biodegradability was excellent for MS-AERBO composite which showed 14.28% weight loss in 28 days, and dimensional stability was excellent for MS-AERBO (modified shellac with AERBO) composite, and SEM analysis showed close bonding with MS-AERBO composite. Thermal stability and dynamic mechanical analysis were also excellent with respect to MS-AERBO composite, and an increase in storage and loss modulus and decline in Tan δ were seen with MS-AERBO composite as compared to AERBO and polystyrene which showed lower stability. [ABSTRACT FROM AUTHOR]

In this work, Abelmoschus manihot waste fibre was utilized to fabricate epoxy-based composite; raw, 15% alkali-treated and 130% grafted AMF composites were prepared with 10%, 20% and 30% loadings, and the mechanical properties were evaluated. It was observed that, at 30% loading, 130% grafted composite gave the highest tensile strength of 65.73 MPa and the flexural, impact strength, interlaminar shear strength and microhardness gave 130%, 7.06 J, 67.9 MPa and 64%, respectively. Chemical resistance was also found to be excellent for grafted AMF at 30% loading. Similarly, the DMA analysis indicated an increase in storage and loss modulus with grafted fibre composite and a decrease in Tan δ with 30% loadings. Dimensional stability was excellent with respect to grafted AMF composites. SEM analysis proved superior bonding with grafted AMF composites followed by alkali and raw composites. The vertical burning test took a long time to burn off for the grafted AMF composites at 30% loading. TGA proved to be excellent with higher loading in the case of grafted AMF composite. Void volume decreased with respect to fibre loading and with the help of surface modification using alkali treatment and grafting of the fibres. [ABSTRACT FROM AUTHOR]

This paper deals with the graft copolymerization of methyl methacrylate (MMA) on the backbone of Abelmoschus manihot fibres (AMF) successfully carried out under aqueous medium using Fenton's reagent [(ferrous ammonium sulphate (FAS) (8.0 × 10–3 M) and H2O2 (0.12 M)] as a redox initiator and N,N′-methylene bisacrylamide as a crosslinker to impart hydrophobic properties. Grafting was optimized under different conditions concerning time, temperature, monomer ratio, material to liquor ratio, initiator concentration. Maximum graft add-on% obtained was 130% at a temperature of 75 °C, 3:1 monomer to fibre ratio, time of 150 min, 1:75 fibre to liquor ratio and 20:20 (FAS: H2O2) initiator concentration. The grafted AMF fibres were characterized using Fourier transform infrared (FT-IR) spectroscopy, scanning electron microscopy (SEM), thermogravimetric analysis (TGA) and X-ray diffraction (XRD) studies. FT-IR spectroscopic data showed a peak at 1726 cm−1 of C=O (ester), and another peak at 1650 cm−1 of adsorbed water was absent in the grafted fibre. The grafted fibre showed excellent thermal stability with an overall weight loss of 51.8% and char residue of 48.2% as well as X-ray diffraction of MMA grafted AMF showed 23% crystallinity and crystallinity index of 0.19. These raw and grafted fibres also subjected to physico-chemical properties like moisture absorbance, swelling behaviour and chemical resistance studies showed that MMA grafted AMF exhibited hydrophobic properties. Therefore, due to this hydrophobic behaviour, the oil absorbency of MMA grafted AMF was studied towards four emulsions: crude oil saline, engine oil saline, diesel oil saline and used engine oil saline. In total, 130% grafted AMF fibre absorbed 30.07 g of crude oil, 12.20 g of diesel oil, 16.00 g of engine oil and 17.32 g of used engine oil. These results showed that the oil removal efficiency of the grafted AMF was intact by oil fraction type and can be applied for all kinds of oil fraction spillage. [ABSTRACT FROM AUTHOR]

With the increased interest in smart cities and smart infrastructures, the need for energy conservation is increasing. Especially with the current electrical energy production mainly relying on nonrenewable resources, conservation of electrical energy is one of the challenging aspects across the globe. However, one can only perform energy conservation optimally by identifying consumption patterns at a granular level, which requires accurate and ubiquitous monitoring infrastructure. Because the electrical energy wastage can occur at any granularity (from a small house-hold appliance to grid-level wastage), the development of a low-cost, easy-to-install, and accurate power-monitoring infrastructure is need of the hour. Hence, in this article, we propose the developed designs for IoT-enabled power monitoring. First is the noninvasive power monitor with voltage connection. The second design introduces a novel split architecture with centralized voltage measurement, which removes the need for local voltage measurements. We have proposed the third and final version of the IoT-enabled power monitor to fulfill the need for three-phase power monitoring. Unlike first and second designs, this design can be used with noninvasive and invasive current sensors. The proposed architecture also supports essential features, such as secure data transfer. Developed devices transmit real-time data to the cloud server, which makes the data ubiquitously available anywhere and anytime. For analyzing the performance of the proposed architecture, the developed devices are deployed in real industrial scenarios. As an example use case, the electrical anomaly detection framework using the data collected is also explained, and the corresponding results are discussed. [ABSTRACT FROM AUTHOR]